I am applying for funding of a K07 Preventive Oncology Training Grant that will facilitate my entry into a laboratory-based translational research career in cancer prevention and diagnostic biomarkers. As an academic pathologist, I have expertise in tissue based assays and tissue microdissection, but am seeking training in high throughput molecular diagnostic techniques (e.g. array based genomic chromosomal hybridization {array CGH)) combined with informatics, biostatistics, epidemiology and cancer prevention clinical trials design. In this proposal, I outline needed graduate coursework in these research disciplines that will enable my completion of a hypothesis-driven research plan to analyze genomic alteration patterns associated with colorectal adenoma recurrence and high risk adenomas. My primary mentor is Dr. Dave Alberts, principal investigator on large Phase III clinical trials on chemoprevention of colorectal adenoma recurrence. Dr. Martinez offers mentorship in epidemiology, Drs. Trent and Bittner offer expertise in genome screening techniques, and Drs. Nagle and Bhattacharyya offers expertise in validation schemes for phenotypic correlates. Graduate coursework in statistics and a strong team of informatics consultants contribute to my training in data analysis and data mining strategies. Hypothesis: Analysis of colorectal adenomas by array CGH will generate molecular profiles predictive of adenoma recurrence and of high risk for neoplastic progression. Specific Aims: 1. To perform genome-wide analysis by array CGH using formalin fixed paraffin embedded clinical trials material. 2. To generate composite molecular profiles by array CGH predictive of adenoma recurrence by analysis of matched baseline and recurrent adenomas. 3. To analyze phenotypic advanced adenomas to generate composite profiles with high risk of neoplastic progression. Patient Population: Two large phase III clinical trials assessing diet (Wheat Bran Fiber) and chemoprevention (Ursodeoxycholic Acid) effects on adenoma recurrence with over 2800 patients accrued and recurrence rates of 40 to 50% at 3-year follow-up colonoscopy. Tissue fixation: 10% buffered formalin. Microarray: Agilent 60-mer oligonucleotide microarrays and image scanner. Sample size: 50 matched pairs, baseline and recurrent; validation studies, tissue microarrays, FISH and RT-PCR. Bioinformatics/Biostatistics: Nonsupervised data analysis, classifier algorithms, including strong features.